The Lawyer in Your Head:
Why Smart People Miss Their Own Bias

Back to Articles

Between January and September of 1998, the most credentialed fund on Wall Street lost more than ninety percent of its capital.

Long-Term Capital Management had Myron Scholes and Robert Merton on its board. The two of them had shared the Nobel in economics the year before, for the formula traders still use to price options. Its founder, John Meriwether, had run the bond desk at Salomon Brothers. The fund started trading in February 1994 with just over a billion dollars, and by the Federal Reserve's own account it returned 20 percent in 1994, 43 in 1995, 41 in 1996 and 17 in 1997.

Then Russia devalued its currency and stopped paying its debts. Spreads that were supposed to converge went the other way, nearly all of them at once, and the borrowed money that had made the good years so good started running in reverse. The fund lost 44 percent of its value in August alone. On September 28, fourteen banks put about $3.6 billion into it, in a rescue the Federal Reserve Bank of New York organized because letting that particular fund fail looked like it might take other firms down with it.

Nobody at LTCM misplaced a decimal. The math was mostly right. What broke was the people wrapped around the math: too much borrowed money, and too much confidence that markets would keep behaving the way the models said they should. Being right eventually is a different thing from being right on time.

The test you think you're taking

Investing looks like an IQ test. Charts, ratios, and a whole industry dressed in the costume of precision. School spent sixteen years teaching us that the sharpest person in the room wins, so we assume money keeps score the same way.

Some of that is commercial. Research can be sold. Forecasts, screeners, a faster feed, a smarter model, all of it can be packaged and priced. Nobody can charge you much for sitting still.

Which is fine as far as it goes. The trouble is the second assumption underneath it: that if you're smart enough, you'll notice when you're the problem.

The blind spot

In 2012, three researchers published a paper in the Journal of Personality and Social Psychology with a title that gives away the ending: "Cognitive Sophistication Does Not Attenuate the Bias Blind Spot."

Richard West, Russell Meserve and Keith Stanovich ran people through the classic thinking traps, the ones Kahneman and Tversky catalogued decades earlier. Then they asked a second question. Compared with the average person, how prone to these errors are you? Nearly everyone gave the same answer. Less prone than average.

That distance, between how easily we spot the error in someone else and how invisible it is in ourselves, is the bias blind spot. What the paper set out to test was whether being smarter closes it.

It doesn't. In the authors' own words: "If anything, a larger bias blind spot was associated with higher cognitive ability."

I think the mechanism is ordinary. A sharp mind is a very good lawyer. Hand it a verdict you've already reached, something like I should get out before this gets worse, and it'll build you a beautiful case, with exhibits. More horsepower buys you a better argued brief for the decision you were always going to make. You've felt this. The nights you researched hardest were probably the nights you'd already decided.

What it costs

Morningstar runs a study every year called Mind the Gap. The 2025 edition compared what funds earned against what the people inside those funds actually earned, over the ten years ending December 31, 2024.

The funds returned 8.2 percent a year. Their investors earned 7.0 percent. Same funds.

That 1.2 point difference is timing: money showing up after the good stretch and leaving after the bad one. Across the decade it comes to roughly 15 percent of the total return those funds produced, given away by the people who owned them. Morningstar also sorted funds by how much money moved in and out. The calmest fifth gave up 0.8 points a year. The busiest fifth gave up 1.8.

Morningstar is careful about how it reads this, and so am I. Some of the gap is just ordinary life, money arriving with a paycheck instead of in one lump at the start, which is a good habit that still opens a gap. But the pattern held in every version they cut. The more the money moved, the less of it was left.

What the fix actually looks like

Up close, temperament looks less like character and more like paperwork. Arrangements made on a calm day that don't ask how you feel later.

Contributions that leave the paycheck on their own, in good months and ugly ones. An allocation chosen when nothing is on fire and written down, so the future panicked version of you gets instructions from the calm version. Fewer logins. I've come to believe that checking your account is one of the most underrated risks in investing, because every login is a fresh invitation to do something.

None of that needs a finance degree. An automatic transfer takes about ten minutes to set up. Deciding an allocation once, on paper, takes an evening. Small mechanical moves, available to anyone, and the Morningstar gap is a rough measure of what they're worth.

Most weeks I'm sitting in conversations about retirement money, and one pattern took me a while to accept. The calmest people in those conversations are rarely the ones who know the most about markets. Some of the sharpest are one headline away from undoing ten years of good decisions. What separates them isn't what they know. It's what they no longer have to decide.

LTCM had a billion dollars, two Nobel laureates and the best known bond trader alive. It lasted four and a half years.

Somewhere out there is a saver who set up a payroll deduction in 1994 and never did anything clever again. Thirty-two years later they're still in. Nobody will write a book about them, which is more or less the point.